LiteLLM PDF Input Bytes
Pass a downloaded PDF as raw bytes to a LiteLLM agent and summarize its contents.
"""
Litellm Pdf Input Bytes
=======================
Cookbook example for `litellm/pdf_input_bytes.py`.
"""
from pathlib import Path
from agno.agent import Agent
from agno.media import File
from agno.models.litellm import LiteLLM
from agno.utils.media import download_file
# ---------------------------------------------------------------------------
# Create Agent
# ---------------------------------------------------------------------------
pdf_path = Path(__file__).parent.joinpath("ThaiRecipes.pdf")
download_file(
"https://agno-public.s3.amazonaws.com/recipes/ThaiRecipes.pdf", str(pdf_path)
)
agent = Agent(
model=LiteLLM(id="openai/gpt-5.6-luna"),
markdown=True,
)
agent.print_response(
"Summarize the contents of the attached file.",
files=[
File(
content=pdf_path.read_bytes(),
),
],
)
# ---------------------------------------------------------------------------
# Run Agent
# ---------------------------------------------------------------------------
if __name__ == "__main__":
passRun the Example
Set up your virtual environment
uv venv --python 3.12
source .venv/bin/activateInstall dependencies
uv pip install -U agno litellmSet your OpenAI credentials
Use an OpenAI API key with access to the requested model. The LiteLLM SDK calls the provider directly. An existing LITELLM_API_KEY overrides provider-specific credentials, so clear it for this example.
unset LITELLM_API_KEY
export OPENAI_API_KEY="your_provider_api_key_here"Set compatible sampling options
Add temperature=None, top_p=None to every LiteLLM(...) using id="gpt-5.6-luna" or id="openai/gpt-5.6-luna" in your saved file. The adapter defaults to temperature=0.7 and top_p=1.0; the LiteLLM SDK rejects those sampling settings for this model's default reasoning mode before sending a request.
Run the example
Save the code above as pdf_input_bytes.py, then run:
python pdf_input_bytes.pyFull source: cookbook/90_models/litellm/pdf_input_bytes.py